Abstract Tuberculosis (TB) is caused by Mycobaterium tuberculosis (Mtb). It is estimated that 2 billion people worldwide are infected, with most infections being latent and asymptomatic and most people living in developing countries. Effective treatment and control of TB requires a diagnostic test that has three critical characteristics: it should (i) detect early-stage disease, (ii) differentially identify latent versus active infections and (iii) utilize methods that are rapid, inexpensive and simple-to-use. The most used diagnostic tests, the tuberculin skin test (TST) and interferon-gamma release assay (IGRA) blood test, do not possess these characteristics. These tests do not detect early-stage infections when symptoms are non-obvious, bacterial numbers are low, and the disease is most controllable, and do not provide definitive differential diagnosis among active infection, latent infection and cleared infection or BCG vaccination. These assays measure immune responses as indirect measures of disease and thus often do not reflect current Mtb infection status. These tests require expert healthcare personnel, can be costly (IGRA) and can take 2 ? 3 days to perform and require two visits (TST). In response, we developed a Luminex fluorescent bead-based test (Magpix) to detect early-stage disease that distinguishes between active and latent infections by directly measuring concentrations of proteins secreted by viable Mtb bacilli: alpha-crytallin and early-secretory antigenic target-6. More recently, we developed microneedle patches (MNPs) that are inexpensive to manufacture, are simple to use and enable rapid analysis. MNPs are skin patches containing micron-scale needles that can be coated with bioactive molecules including antibodies. In this project, we will optimize a MNP-based diagnostic test which has all three critical characteristics described above that are needed as a first step towards effective treatment and control of TB. Using this approach, minimally trained personnel can apply a MNP to the skin of a patient for a few minutes. During that time, Mtb-secreted proteins in dermal interstitial fluid (ISF) are collected and subsequently assayed. Aim 1 of the proposed study will develop MNPs to assay Mtb antigens as a TB diagnostic platform using two different approaches. First, MNPs will be used to collect ISF from the skin, which will subsequently be assayed for Mtb-secreted proteins by sandwich ELISA. A second approach attaches antibodies onto MNPs to capture Mtb proteins in the skin, which are then analyzed in situ on the MNP by immunoassay. Each MNP design will be optimized for assay sensitivity and specificity using Mtb proteins in spiked guinea pig ISF. In Aim 2, the best-performing MNP design will be tested on skin of Mtb-infected guinea pigs for diagnostic efficacy, as a monitor for therapeutic efficacy, safety, ease of use, and robustness. Future studies will assess assay performance in humans in a TB-endemic area for diagnosis of infection during all stages of the disease and accurate assessments of treatment efficacy.